the root cause is something blocking the DB lock... we are not sure what.
The consequence is that ods-ksmutil processes wait forever while they retry getting the lock themselves.
The fix is to only retry for 1 minute then give up and die, a patch for 1.3 has been sent to cira for them to test.